1. Field of the Invention
The present invention relates to a decoding method and a memory managing method, and more particularly, to a video data decoding method and a memory managing method for video data decoding process.
2. Description of the Related Art
Generally speaking, the amount of the video data is usually enormous. In order to save the space for storing the video data, and to save the transmission bandwidth for transmitting the video data, the video data have to be compressed before it is transmitted. The object of the data compression is achieved by removing the superfluous information from the video data. For example, if a subsequent frame is similar to the previous frame, the previous frame is preserved and the same content in the subsequent frames are removed (that is, only the information with different content is preserved), such that the amount of the digital video data is significantly reduced.
For the MPEG (Motion Picture Experts Group) standard, the video decoding process usually produces three types of frames, which are intra-coding frame (I frame), prediction frame (P frame), and Bi-directional frame (B frame). It should be noted that P frame is decoded with reference to information of I frame stored in the memory. In addition, B frame is decoded with reference to information of both I frame and P frame stored in the memory.
When the High Definition Television (HDTV) or Digital Video Disc (DVD) data are being decoded, the memory size for storing I frame, P frame, and B frame are increased according to the big size of the video data. For example, in the specification defined by the Advanced Television Systems Committee (ATSC), the memory size and bandwidth for 1920×1080 interleaved displaying is about 6 times the memory size of the specification defined by the National Television System Committee (NTSC).
In a system having a limited memory size, if the data amount of the reference frame is directly reduced without considering other factors, the quality of the frame decoded based on the reference frame would deteriorate, which causes significant degradation of the group of picture quality. In addition, in order to avoid the significant degradation of the group of picture quality, the memory size has to be increased, which inevitably increases the manufacturing cost.